2 * Jalview - A Sequence Alignment Editor and Viewer ($$Version-Rel$$)
3 * Copyright (C) $$Year-Rel$$ The Jalview Authors
5 * This file is part of Jalview.
7 * Jalview is free software: you can redistribute it and/or
8 * modify it under the terms of the GNU General Public License
9 * as published by the Free Software Foundation, either version 3
10 * of the License, or (at your option) any later version.
12 * Jalview is distributed in the hope that it will be useful, but
13 * WITHOUT ANY WARRANTY; without even the implied warranty
14 * of MERCHANTABILITY or FITNESS FOR A PARTICULAR
15 * PURPOSE. See the GNU General Public License for more details.
17 * You should have received a copy of the GNU General Public License
18 * along with Jalview. If not, see <http://www.gnu.org/licenses/>.
19 * The Jalview Authors are detailed in the 'AUTHORS' file.
21 package jalview.datamodel;
23 import jalview.analysis.AlignSeq;
24 import jalview.api.DBRefEntryI;
25 import jalview.datamodel.features.SequenceFeatures;
26 import jalview.datamodel.features.SequenceFeaturesI;
27 import jalview.util.Comparison;
28 import jalview.util.DBRefUtils;
29 import jalview.util.MapList;
30 import jalview.util.StringUtils;
32 import java.util.ArrayList;
33 import java.util.Arrays;
34 import java.util.BitSet;
35 import java.util.Collections;
36 import java.util.Enumeration;
37 import java.util.List;
38 import java.util.ListIterator;
39 import java.util.Vector;
41 import com.stevesoft.pat.Regex;
43 import fr.orsay.lri.varna.models.rna.RNA;
47 * Implements the SequenceI interface for a char[] based sequence object.
52 public class Sequence extends ASequence implements SequenceI
54 private static final Regex limitrx = new Regex(
55 "[/][0-9]{1,}[-][0-9]{1,}$");
57 private static final Regex endrx = new Regex("[0-9]{1,}$");
59 SequenceI datasetSequence;
63 private char[] sequence;
71 Vector<PDBEntry> pdbIds;
80 * This annotation is displayed below the alignment but the positions are tied
81 * to the residues of this sequence
83 * TODO: change to List<>
85 Vector<AlignmentAnnotation> annotation;
88 * The index of the sequence in a MSA
92 private SequenceFeaturesI sequenceFeatureStore;
95 * A cursor holding the approximate current view position to the sequence,
96 * as determined by findIndex or findPosition or findPositions.
97 * Using a cursor as a hint allows these methods to be more performant for
100 private SequenceCursor cursor;
103 * A number that should be incremented whenever the sequence is edited.
104 * If the value matches the cursor token, then we can trust the cursor,
105 * if not then it should be recomputed.
107 private int changeCount;
110 * Creates a new Sequence object.
113 * display name string
115 * string to form a possibly gapped sequence out of
117 * first position of non-gap residue in the sequence
119 * last position of ungapped residues (nearly always only used for
122 public Sequence(String name, String sequence, int start, int end)
125 initSeqAndName(name, sequence.toCharArray(), start, end);
128 public Sequence(String name, char[] sequence, int start, int end)
131 initSeqAndName(name, sequence, start, end);
135 * Stage 1 constructor - assign name, sequence, and set start and end fields.
136 * start and end are updated values from name2 if it ends with /start-end
143 protected void initSeqAndName(String name2, char[] sequence2, int start2,
147 this.sequence = sequence2;
159 "POSSIBLE IMPLEMENTATION ERROR: null sequence name passed to constructor.");
162 // Does sequence have the /start-end signature?
163 if (limitrx.search(name))
165 name = limitrx.left();
166 endrx.search(limitrx.stringMatched());
167 setStart(Integer.parseInt(limitrx.stringMatched().substring(1,
168 endrx.matchedFrom() - 1)));
169 setEnd(Integer.parseInt(endrx.stringMatched()));
173 void checkValidRange()
176 // http://issues.jalview.org/browse/JAL-774?focusedCommentId=11239&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-11239
179 for (int j = 0; j < sequence.length; j++)
181 if (!jalview.util.Comparison.isGap(sequence[j]))
200 * default constructor
204 sequenceFeatureStore = new SequenceFeatures();
208 * Creates a new Sequence object.
215 public Sequence(String name, String sequence)
217 this(name, sequence, 1, -1);
221 * Creates a new Sequence object with new AlignmentAnnotations but inherits
222 * any existing dataset sequence reference. If non exists, everything is
226 * if seq is a dataset sequence, behaves like a plain old copy
229 public Sequence(SequenceI seq)
231 this(seq, seq.getAnnotation());
235 * Create a new sequence object with new features, DBRefEntries, and PDBIds
236 * but inherits any existing dataset sequence reference, and duplicate of any
237 * annotation that is present in the given annotation array.
240 * the sequence to be copied
241 * @param alAnnotation
242 * an array of annotation including some associated with seq
244 public Sequence(SequenceI seq, AlignmentAnnotation[] alAnnotation)
247 initSeqFrom(seq, alAnnotation);
251 * does the heavy lifting when cloning a dataset sequence, or coping data from
252 * dataset to a new derived sequence.
255 * - source of attributes.
256 * @param alAnnotation
257 * - alignment annotation present on seq that should be copied onto
260 protected void initSeqFrom(SequenceI seq,
261 AlignmentAnnotation[] alAnnotation)
263 char[] oseq = seq.getSequence(); // returns a copy of the array
264 initSeqAndName(seq.getName(), oseq, seq.getStart(), seq.getEnd());
266 description = seq.getDescription();
267 if (seq != datasetSequence)
269 setDatasetSequence(seq.getDatasetSequence());
273 * only copy DBRefs and seqfeatures if we really are a dataset sequence
275 if (datasetSequence == null)
277 if (seq.getDBRefs() != null)
279 DBRefEntry[] dbr = seq.getDBRefs();
280 for (int i = 0; i < dbr.length; i++)
282 addDBRef(new DBRefEntry(dbr[i]));
287 * make copies of any sequence features
289 for (SequenceFeature sf : seq.getSequenceFeatures())
291 addSequenceFeature(new SequenceFeature(sf));
295 if (seq.getAnnotation() != null)
297 AlignmentAnnotation[] sqann = seq.getAnnotation();
298 for (int i = 0; i < sqann.length; i++)
300 if (sqann[i] == null)
304 boolean found = (alAnnotation == null);
307 for (int apos = 0; !found && apos < alAnnotation.length; apos++)
309 found = (alAnnotation[apos] == sqann[i]);
314 // only copy the given annotation
315 AlignmentAnnotation newann = new AlignmentAnnotation(sqann[i]);
316 addAlignmentAnnotation(newann);
320 if (seq.getAllPDBEntries() != null)
322 Vector<PDBEntry> ids = seq.getAllPDBEntries();
323 for (PDBEntry pdb : ids)
325 this.addPDBId(new PDBEntry(pdb));
331 public void setSequenceFeatures(List<SequenceFeature> features)
333 if (datasetSequence != null)
335 datasetSequence.setSequenceFeatures(features);
338 sequenceFeatureStore = new SequenceFeatures(features);
342 public synchronized boolean addSequenceFeature(SequenceFeature sf)
344 if (sf.getType() == null)
346 System.err.println("SequenceFeature type may not be null: "
351 if (datasetSequence != null)
353 return datasetSequence.addSequenceFeature(sf);
356 return sequenceFeatureStore.add(sf);
360 public void deleteFeature(SequenceFeature sf)
362 if (datasetSequence != null)
364 datasetSequence.deleteFeature(sf);
368 sequenceFeatureStore.delete(sf);
378 public List<SequenceFeature> getSequenceFeatures()
380 if (datasetSequence != null)
382 return datasetSequence.getSequenceFeatures();
384 return sequenceFeatureStore.getAllFeatures();
388 public SequenceFeaturesI getFeatures()
390 return datasetSequence != null ? datasetSequence.getFeatures()
391 : sequenceFeatureStore;
395 public boolean addPDBId(PDBEntry entry)
399 pdbIds = new Vector<PDBEntry>();
404 for (PDBEntry pdbe : pdbIds)
406 if (pdbe.updateFrom(entry))
411 pdbIds.addElement(entry);
422 public void setPDBId(Vector<PDBEntry> id)
430 * @return DOCUMENT ME!
433 public Vector<PDBEntry> getAllPDBEntries()
435 return pdbIds == null ? new Vector<PDBEntry>() : pdbIds;
441 * @return DOCUMENT ME!
444 public String getDisplayId(boolean jvsuffix)
446 StringBuffer result = new StringBuffer(name);
449 result.append("/" + start + "-" + end);
452 return result.toString();
462 public void setName(String name)
471 * @return DOCUMENT ME!
474 public String getName()
486 public void setStart(int start)
494 * @return DOCUMENT ME!
497 public int getStart()
509 public void setEnd(int end)
517 * @return DOCUMENT ME!
528 * @return DOCUMENT ME!
531 public int getLength()
533 return this.sequence.length;
543 public void setSequence(String seq)
545 this.sequence = seq.toCharArray();
551 public String getSequenceAsString()
553 return new String(sequence);
557 public String getSequenceAsString(int start, int end)
559 return new String(getSequence(start, end));
563 public char[] getSequence()
566 return sequence == null ? null : Arrays.copyOf(sequence,
573 * @see jalview.datamodel.SequenceI#getSequence(int, int)
576 public char[] getSequence(int start, int end)
582 // JBPNote - left to user to pad the result here (TODO:Decide on this
584 if (start >= sequence.length)
589 if (end >= sequence.length)
591 end = sequence.length;
594 char[] reply = new char[end - start];
595 System.arraycopy(sequence, start, reply, 0, end - start);
601 public SequenceI getSubSequence(int start, int end)
607 char[] seq = getSequence(start, end);
612 int nstart = findPosition(start);
613 int nend = findPosition(end) - 1;
614 // JBPNote - this is an incomplete copy.
615 SequenceI nseq = new Sequence(this.getName(), seq, nstart, nend);
616 nseq.setDescription(description);
617 if (datasetSequence != null)
619 nseq.setDatasetSequence(datasetSequence);
623 nseq.setDatasetSequence(this);
629 * Returns the character of the aligned sequence at the given position (base
630 * zero), or space if the position is not within the sequence's bounds
635 public char getCharAt(int i)
637 if (i >= 0 && i < sequence.length)
648 * Sets the sequence description, and also parses out any special formats of
654 public void setDescription(String desc)
656 this.description = desc;
660 public void setGeneLoci(String speciesId, String assemblyId,
661 String chromosomeId, MapList map)
663 addDBRef(new DBRefEntry(speciesId, assemblyId, DBRefEntry.CHROMOSOME
664 + ":" + chromosomeId, new Mapping(map)));
668 * Returns the gene loci mapping for the sequence (may be null)
673 public GeneLociI getGeneLoci()
675 DBRefEntry[] refs = getDBRefs();
678 for (final DBRefEntry ref : refs)
680 if (ref.isChromosome())
682 return new GeneLociI()
685 public String getSpeciesId()
687 return ref.getSource();
691 public String getAssemblyId()
693 return ref.getVersion();
697 public String getChromosomeId()
699 // strip of "chromosome:" prefix to chrId
700 return ref.getAccessionId().substring(
701 DBRefEntry.CHROMOSOME.length() + 1);
705 public MapList getMap()
707 return ref.getMap().getMap();
717 * Answers the description
722 public String getDescription()
724 return this.description;
731 public int findIndex(int pos)
734 * use a valid, hopefully nearby, cursor if available
736 if (isValidCursor(cursor))
738 return findIndex(pos, cursor);
746 * traverse sequence from the start counting gaps; make a note of
747 * the column of the first residue to save in the cursor
749 while ((i < sequence.length) && (j <= end) && (j <= pos))
751 if (!Comparison.isGap(sequence[i]))
762 if (j == end && j < pos)
767 updateCursor(pos, i, startColumn);
772 * Updates the cursor to the latest found residue and column position
779 * column position of the first sequence residue
781 protected void updateCursor(int residuePos, int column, int startColumn)
784 * preserve end residue column provided cursor was valid
786 int endColumn = isValidCursor(cursor) ? cursor.lastColumnPosition : 0;
787 if (residuePos == this.end)
792 cursor = new SequenceCursor(this, residuePos, column, startColumn,
793 endColumn, this.changeCount);
797 * Answers the aligned column position (1..) for the given residue position
798 * (start..) given a 'hint' of a residue/column location in the neighbourhood.
799 * The hint may be left of, at, or to the right of the required position.
805 protected int findIndex(int pos, SequenceCursor curs)
807 if (!isValidCursor(curs))
810 * wrong or invalidated cursor, compute de novo
812 return findIndex(pos);
815 if (curs.residuePosition == pos)
817 return curs.columnPosition;
821 * move left or right to find pos from hint.position
823 int col = curs.columnPosition - 1; // convert from base 1 to 0-based array
825 int newPos = curs.residuePosition;
826 int delta = newPos > pos ? -1 : 1;
828 while (newPos != pos)
830 col += delta; // shift one column left or right
831 if (col < 0 || col == sequence.length)
835 if (!Comparison.isGap(sequence[col]))
841 col++; // convert back to base 1
842 updateCursor(pos, col, curs.firstColumnPosition);
851 public int findPosition(final int column)
854 * use a valid, hopefully nearby, cursor if available
856 if (isValidCursor(cursor))
858 return findPosition(column + 1, cursor);
861 // TODO recode this more naturally i.e. count residues only
862 // as they are found, not 'in anticipation'
865 * traverse the sequence counting gaps; note the column position
866 * of the first residue, to save in the cursor
868 int firstResidueColumn = 0;
869 int lastPosFound = 0;
870 int lastPosFoundColumn = 0;
871 int seqlen = sequence.length;
873 if (seqlen > 0 && !Comparison.isGap(sequence[0]))
875 lastPosFound = start;
876 lastPosFoundColumn = 0;
882 while (j < column && j < seqlen)
884 if (!Comparison.isGap(sequence[j]))
887 lastPosFoundColumn = j;
888 if (pos == this.start)
890 firstResidueColumn = j;
896 if (j < seqlen && !Comparison.isGap(sequence[j]))
899 lastPosFoundColumn = j;
900 if (pos == this.start)
902 firstResidueColumn = j;
907 * update the cursor to the last residue position found (if any)
908 * (converting column position to base 1)
910 if (lastPosFound != 0)
912 updateCursor(lastPosFound, lastPosFoundColumn + 1,
913 firstResidueColumn + 1);
920 * Answers true if the given cursor is not null, is for this sequence object,
921 * and has a token value that matches this object's changeCount, else false.
922 * This allows us to ignore a cursor as 'stale' if the sequence has been
923 * modified since the cursor was created.
928 protected boolean isValidCursor(SequenceCursor curs)
930 if (curs == null || curs.sequence != this || curs.token != changeCount)
935 * sanity check against range
937 if (curs.columnPosition < 0 || curs.columnPosition > sequence.length)
941 if (curs.residuePosition < start || curs.residuePosition > end)
949 * Answers the sequence position (start..) for the given aligned column
950 * position (1..), given a hint of a cursor in the neighbourhood. The cursor
951 * may lie left of, at, or to the right of the column position.
957 protected int findPosition(final int col, SequenceCursor curs)
959 if (!isValidCursor(curs))
962 * wrong or invalidated cursor, compute de novo
964 return findPosition(col - 1);// ugh back to base 0
967 if (curs.columnPosition == col)
969 cursor = curs; // in case this method becomes public
970 return curs.residuePosition; // easy case :-)
973 if (curs.lastColumnPosition > 0 && curs.lastColumnPosition < col)
976 * sequence lies entirely to the left of col
977 * - return last residue + 1
982 if (curs.firstColumnPosition > 0 && curs.firstColumnPosition > col)
985 * sequence lies entirely to the right of col
986 * - return first residue
991 // todo could choose closest to col out of column,
992 // firstColumnPosition, lastColumnPosition as a start point
995 * move left or right to find pos from cursor position
997 int firstResidueColumn = curs.firstColumnPosition;
998 int column = curs.columnPosition - 1; // to base 0
999 int newPos = curs.residuePosition;
1000 int delta = curs.columnPosition > col ? -1 : 1;
1001 boolean gapped = false;
1002 int lastFoundPosition = curs.residuePosition;
1003 int lastFoundPositionColumn = curs.columnPosition;
1005 while (column != col - 1)
1007 column += delta; // shift one column left or right
1008 if (column < 0 || column == sequence.length)
1012 gapped = Comparison.isGap(sequence[column]);
1016 lastFoundPosition = newPos;
1017 lastFoundPositionColumn = column + 1;
1018 if (lastFoundPosition == this.start)
1020 firstResidueColumn = column + 1;
1025 if (cursor == null || lastFoundPosition != cursor.residuePosition)
1027 updateCursor(lastFoundPosition, lastFoundPositionColumn,
1028 firstResidueColumn);
1032 * hack to give position to the right if on a gap
1033 * or beyond the length of the sequence (see JAL-2562)
1035 if (delta > 0 && (gapped || column >= sequence.length))
1047 public Range findPositions(int fromColumn, int toColumn)
1049 if (toColumn < fromColumn || fromColumn < 1)
1055 * find the first non-gapped position, if any
1057 int firstPosition = 0;
1058 int col = fromColumn - 1;
1059 int length = sequence.length;
1060 while (col < length && col < toColumn)
1062 if (!Comparison.isGap(sequence[col]))
1064 firstPosition = findPosition(col++);
1070 if (firstPosition == 0)
1076 * find the last non-gapped position
1078 int lastPosition = firstPosition;
1079 while (col < length && col < toColumn)
1081 if (!Comparison.isGap(sequence[col++]))
1087 return new Range(firstPosition, lastPosition);
1091 * Returns an int array where indices correspond to each residue in the
1092 * sequence and the element value gives its position in the alignment
1094 * @return int[SequenceI.getEnd()-SequenceI.getStart()+1] or null if no
1095 * residues in SequenceI object
1098 public int[] gapMap()
1100 String seq = jalview.analysis.AlignSeq.extractGaps(
1101 jalview.util.Comparison.GapChars, new String(sequence));
1102 int[] map = new int[seq.length()];
1106 while (j < sequence.length)
1108 if (!jalview.util.Comparison.isGap(sequence[j]))
1120 public int[] findPositionMap()
1122 int map[] = new int[sequence.length];
1125 int seqlen = sequence.length;
1126 while ((j < seqlen))
1129 if (!jalview.util.Comparison.isGap(sequence[j]))
1140 public List<int[]> getInsertions()
1142 ArrayList<int[]> map = new ArrayList<int[]>();
1143 int lastj = -1, j = 0;
1145 int seqlen = sequence.length;
1146 while ((j < seqlen))
1148 if (jalview.util.Comparison.isGap(sequence[j]))
1159 map.add(new int[] { lastj, j - 1 });
1167 map.add(new int[] { lastj, j - 1 });
1174 public BitSet getInsertionsAsBits()
1176 BitSet map = new BitSet();
1177 int lastj = -1, j = 0;
1179 int seqlen = sequence.length;
1180 while ((j < seqlen))
1182 if (jalview.util.Comparison.isGap(sequence[j]))
1208 public void deleteChars(int i, int j)
1210 int newstart = start, newend = end;
1211 if (i >= sequence.length || i < 0)
1216 char[] tmp = StringUtils.deleteChars(sequence, i, j);
1217 boolean createNewDs = false;
1218 // TODO: take a (second look) at the dataset creation validation method for
1219 // the very large sequence case
1220 int eindex = -1, sindex = -1;
1221 boolean ecalc = false, scalc = false;
1222 for (int s = i; s < j; s++)
1224 if (jalview.schemes.ResidueProperties.aaIndex[sequence[s]] != 23)
1234 sindex = findIndex(start) - 1;
1239 // delete characters including start of sequence
1240 newstart = findPosition(j);
1241 break; // don't need to search for any more residue characters.
1245 // delete characters after start.
1248 eindex = findIndex(end) - 1;
1253 // delete characters at end of sequence
1254 newend = findPosition(i - 1);
1255 break; // don't need to search for any more residue characters.
1260 newend--; // decrease end position by one for the deleted residue
1261 // and search further
1267 // deletion occured in the middle of the sequence
1268 if (createNewDs && this.datasetSequence != null)
1270 // construct a new sequence
1271 Sequence ds = new Sequence(datasetSequence);
1272 // TODO: remove any non-inheritable properties ?
1273 // TODO: create a sequence mapping (since there is a relation here ?)
1274 ds.deleteChars(i, j);
1275 datasetSequence = ds;
1284 public void insertCharAt(int i, int length, char c)
1286 char[] tmp = new char[sequence.length + length];
1288 if (i >= sequence.length)
1290 System.arraycopy(sequence, 0, tmp, 0, sequence.length);
1291 i = sequence.length;
1295 System.arraycopy(sequence, 0, tmp, 0, i);
1305 if (i < sequence.length)
1307 System.arraycopy(sequence, i, tmp, index, sequence.length - i);
1315 public void insertCharAt(int i, char c)
1317 insertCharAt(i, 1, c);
1321 public String getVamsasId()
1327 public void setVamsasId(String id)
1333 public void setDBRefs(DBRefEntry[] dbref)
1335 if (dbrefs == null && datasetSequence != null
1336 && this != datasetSequence)
1338 datasetSequence.setDBRefs(dbref);
1344 DBRefUtils.ensurePrimaries(this);
1349 public DBRefEntry[] getDBRefs()
1351 if (dbrefs == null && datasetSequence != null
1352 && this != datasetSequence)
1354 return datasetSequence.getDBRefs();
1360 public void addDBRef(DBRefEntry entry)
1362 if (datasetSequence != null)
1364 datasetSequence.addDBRef(entry);
1370 dbrefs = new DBRefEntry[0];
1373 for (DBRefEntryI dbr : dbrefs)
1375 if (dbr.updateFrom(entry))
1378 * found a dbref that either matched, or could be
1379 * updated from, the new entry - no need to add it
1386 * extend the array to make room for one more
1388 // TODO use an ArrayList instead
1389 int j = dbrefs.length;
1390 DBRefEntry[] temp = new DBRefEntry[j + 1];
1391 System.arraycopy(dbrefs, 0, temp, 0, j);
1392 temp[temp.length - 1] = entry;
1396 DBRefUtils.ensurePrimaries(this);
1400 public void setDatasetSequence(SequenceI seq)
1404 throw new IllegalArgumentException(
1405 "Implementation Error: self reference passed to SequenceI.setDatasetSequence");
1407 if (seq != null && seq.getDatasetSequence() != null)
1409 throw new IllegalArgumentException(
1410 "Implementation error: cascading dataset sequences are not allowed.");
1412 datasetSequence = seq;
1416 public SequenceI getDatasetSequence()
1418 return datasetSequence;
1422 public AlignmentAnnotation[] getAnnotation()
1424 return annotation == null ? null
1426 .toArray(new AlignmentAnnotation[annotation.size()]);
1430 public boolean hasAnnotation(AlignmentAnnotation ann)
1432 return annotation == null ? false : annotation.contains(ann);
1436 public void addAlignmentAnnotation(AlignmentAnnotation annotation)
1438 if (this.annotation == null)
1440 this.annotation = new Vector<AlignmentAnnotation>();
1442 if (!this.annotation.contains(annotation))
1444 this.annotation.addElement(annotation);
1446 annotation.setSequenceRef(this);
1450 public void removeAlignmentAnnotation(AlignmentAnnotation annotation)
1452 if (this.annotation != null)
1454 this.annotation.removeElement(annotation);
1455 if (this.annotation.size() == 0)
1457 this.annotation = null;
1463 * test if this is a valid candidate for another sequence's dataset sequence.
1466 private boolean isValidDatasetSequence()
1468 if (datasetSequence != null)
1472 for (int i = 0; i < sequence.length; i++)
1474 if (jalview.util.Comparison.isGap(sequence[i]))
1483 public SequenceI deriveSequence()
1485 Sequence seq = null;
1486 if (datasetSequence == null)
1488 if (isValidDatasetSequence())
1490 // Use this as dataset sequence
1491 seq = new Sequence(getName(), "", 1, -1);
1492 seq.setDatasetSequence(this);
1493 seq.initSeqFrom(this, getAnnotation());
1498 // Create a new, valid dataset sequence
1499 createDatasetSequence();
1502 return new Sequence(this);
1505 private boolean _isNa;
1507 private int _seqhash = 0;
1510 * Answers false if the sequence is more than 85% nucleotide (ACGTU), else
1514 public boolean isProtein()
1516 if (datasetSequence != null)
1518 return datasetSequence.isProtein();
1520 if (_seqhash != sequence.hashCode())
1522 _seqhash = sequence.hashCode();
1523 _isNa = Comparison.isNucleotide(this);
1531 * @see jalview.datamodel.SequenceI#createDatasetSequence()
1534 public SequenceI createDatasetSequence()
1536 if (datasetSequence == null)
1538 Sequence dsseq = new Sequence(getName(),
1539 AlignSeq.extractGaps(jalview.util.Comparison.GapChars,
1540 getSequenceAsString()),
1541 getStart(), getEnd());
1543 datasetSequence = dsseq;
1545 dsseq.setDescription(description);
1546 // move features and database references onto dataset sequence
1547 dsseq.sequenceFeatureStore = sequenceFeatureStore;
1548 sequenceFeatureStore = null;
1549 dsseq.dbrefs = dbrefs;
1551 // TODO: search and replace any references to this sequence with
1552 // references to the dataset sequence in Mappings on dbref
1553 dsseq.pdbIds = pdbIds;
1555 datasetSequence.updatePDBIds();
1556 if (annotation != null)
1558 // annotation is cloned rather than moved, to preserve what's currently
1560 for (AlignmentAnnotation aa : annotation)
1562 AlignmentAnnotation _aa = new AlignmentAnnotation(aa);
1563 _aa.sequenceRef = datasetSequence;
1564 _aa.adjustForAlignment(); // uses annotation's own record of
1565 // sequence-column mapping
1566 datasetSequence.addAlignmentAnnotation(_aa);
1570 return datasetSequence;
1577 * jalview.datamodel.SequenceI#setAlignmentAnnotation(AlignmmentAnnotation[]
1581 public void setAlignmentAnnotation(AlignmentAnnotation[] annotations)
1583 if (annotation != null)
1585 annotation.removeAllElements();
1587 if (annotations != null)
1589 for (int i = 0; i < annotations.length; i++)
1591 if (annotations[i] != null)
1593 addAlignmentAnnotation(annotations[i]);
1600 public AlignmentAnnotation[] getAnnotation(String label)
1602 if (annotation == null || annotation.size() == 0)
1607 Vector<AlignmentAnnotation> subset = new Vector<AlignmentAnnotation>();
1608 Enumeration<AlignmentAnnotation> e = annotation.elements();
1609 while (e.hasMoreElements())
1611 AlignmentAnnotation ann = e.nextElement();
1612 if (ann.label != null && ann.label.equals(label))
1614 subset.addElement(ann);
1617 if (subset.size() == 0)
1621 AlignmentAnnotation[] anns = new AlignmentAnnotation[subset.size()];
1623 e = subset.elements();
1624 while (e.hasMoreElements())
1626 anns[i++] = e.nextElement();
1628 subset.removeAllElements();
1633 public boolean updatePDBIds()
1635 if (datasetSequence != null)
1637 // TODO: could merge DBRefs
1638 return datasetSequence.updatePDBIds();
1640 if (dbrefs == null || dbrefs.length == 0)
1644 boolean added = false;
1645 for (DBRefEntry dbr : dbrefs)
1647 if (DBRefSource.PDB.equals(dbr.getSource()))
1650 * 'Add' any PDB dbrefs as a PDBEntry - add is only performed if the
1651 * PDB id is not already present in a 'matching' PDBEntry
1652 * Constructor parses out a chain code if appended to the accession id
1653 * (a fudge used to 'store' the chain code in the DBRef)
1655 PDBEntry pdbe = new PDBEntry(dbr);
1656 added |= addPDBId(pdbe);
1663 public void transferAnnotation(SequenceI entry, Mapping mp)
1665 if (datasetSequence != null)
1667 datasetSequence.transferAnnotation(entry, mp);
1670 if (entry.getDatasetSequence() != null)
1672 transferAnnotation(entry.getDatasetSequence(), mp);
1675 // transfer any new features from entry onto sequence
1676 if (entry.getSequenceFeatures() != null)
1679 List<SequenceFeature> sfs = entry.getSequenceFeatures();
1680 for (SequenceFeature feature : sfs)
1682 SequenceFeature sf[] = (mp != null) ? mp.locateFeature(feature)
1683 : new SequenceFeature[] { new SequenceFeature(feature) };
1686 for (int sfi = 0; sfi < sf.length; sfi++)
1688 addSequenceFeature(sf[sfi]);
1694 // transfer PDB entries
1695 if (entry.getAllPDBEntries() != null)
1697 Enumeration<PDBEntry> e = entry.getAllPDBEntries().elements();
1698 while (e.hasMoreElements())
1700 PDBEntry pdb = e.nextElement();
1704 // transfer database references
1705 DBRefEntry[] entryRefs = entry.getDBRefs();
1706 if (entryRefs != null)
1708 for (int r = 0; r < entryRefs.length; r++)
1710 DBRefEntry newref = new DBRefEntry(entryRefs[r]);
1711 if (newref.getMap() != null && mp != null)
1713 // remap ref using our local mapping
1715 // we also assume all version string setting is done by dbSourceProxy
1717 * if (!newref.getSource().equalsIgnoreCase(dbSource)) {
1718 * newref.setSource(dbSource); }
1726 * @return The index (zero-based) on this sequence in the MSA. It returns
1727 * {@code -1} if this information is not available.
1730 public int getIndex()
1736 * Defines the position of this sequence in the MSA. Use the value {@code -1}
1737 * if this information is undefined.
1740 * position for this sequence. This value is zero-based (zero for
1741 * this first sequence)
1744 public void setIndex(int value)
1750 public void setRNA(RNA r)
1762 public List<AlignmentAnnotation> getAlignmentAnnotations(String calcId,
1765 List<AlignmentAnnotation> result = new ArrayList<AlignmentAnnotation>();
1766 if (this.annotation != null)
1768 for (AlignmentAnnotation ann : annotation)
1770 if (ann.calcId != null && ann.calcId.equals(calcId)
1771 && ann.label != null && ann.label.equals(label))
1781 public String toString()
1783 return getDisplayId(false);
1787 public PDBEntry getPDBEntry(String pdbIdStr)
1789 if (getDatasetSequence() != null)
1791 return getDatasetSequence().getPDBEntry(pdbIdStr);
1797 List<PDBEntry> entries = getAllPDBEntries();
1798 for (PDBEntry entry : entries)
1800 if (entry.getId().equalsIgnoreCase(pdbIdStr))
1809 public List<DBRefEntry> getPrimaryDBRefs()
1811 if (datasetSequence != null)
1813 return datasetSequence.getPrimaryDBRefs();
1815 if (dbrefs == null || dbrefs.length == 0)
1817 return Collections.emptyList();
1819 synchronized (dbrefs)
1821 List<DBRefEntry> primaries = new ArrayList<DBRefEntry>();
1822 DBRefEntry[] tmp = new DBRefEntry[1];
1823 for (DBRefEntry ref : dbrefs)
1825 if (!ref.isPrimaryCandidate())
1831 MapList mp = ref.getMap().getMap();
1832 if (mp.getFromLowest() > start || mp.getFromHighest() < end)
1834 // map only involves a subsequence, so cannot be primary
1838 // whilst it looks like it is a primary ref, we also sanity check type
1839 if (DBRefUtils.getCanonicalName(DBRefSource.PDB)
1840 .equals(DBRefUtils.getCanonicalName(ref.getSource())))
1842 // PDB dbrefs imply there should be a PDBEntry associated
1843 // TODO: tighten PDB dbrefs
1844 // formally imply Jalview has actually downloaded and
1845 // parsed the pdb file. That means there should be a cached file
1846 // handle on the PDBEntry, and a real mapping between sequence and
1847 // extracted sequence from PDB file
1848 PDBEntry pdbentry = getPDBEntry(ref.getAccessionId());
1849 if (pdbentry != null && pdbentry.getFile() != null)
1855 // check standard protein or dna sources
1857 DBRefEntry[] res = DBRefUtils.selectDbRefs(!isProtein(), tmp);
1858 if (res != null && res[0] == tmp[0])
1872 public List<SequenceFeature> findFeatures(int fromColumn, int toColumn,
1875 int startPos = findPosition(fromColumn - 1); // convert base 1 to base 0
1876 int endPos = fromColumn == toColumn ? startPos
1877 : findPosition(toColumn - 1);
1879 List<SequenceFeature> result = getFeatures().findFeatures(startPos,
1883 * if end column is gapped, endPos may be to the right,
1884 * and we may have included adjacent or enclosing features;
1885 * remove any that are not enclosing, non-contact features
1887 if (endPos > this.end || Comparison.isGap(sequence[toColumn - 1]))
1889 ListIterator<SequenceFeature> it = result.listIterator();
1890 while (it.hasNext())
1892 SequenceFeature sf = it.next();
1893 int sfBegin = sf.getBegin();
1894 int sfEnd = sf.getEnd();
1895 int featureStartColumn = findIndex(sfBegin);
1896 if (featureStartColumn > toColumn)
1900 else if (featureStartColumn < fromColumn)
1902 int featureEndColumn = sfEnd == sfBegin ? featureStartColumn
1904 if (featureEndColumn < fromColumn)
1908 else if (featureEndColumn > toColumn && sf.isContactFeature())
1911 * remove an enclosing feature if it is a contact feature
1923 * Invalidates any stale cursors (forcing recalculation) by incrementing the
1924 * token that has to match the one presented by the cursor
1927 public void sequenceChanged()
1936 public int replace(char c1, char c2)
1943 synchronized (sequence)
1945 for (int c = 0; c < sequence.length; c++)
1947 if (sequence[c] == c1)